- Email: [email protected]
Sequence Elements Required for Apolipoprotein B mRNA Editing Enhancement Activity from Chicken Enterocytes
Biochemical and Biophysical Research Communications 254, 744 –750 (1999) Article ID bbrc.1998.9963, available online at http://www.idealibrary.com on
Sequence Elements Required for Apolipoprotein B mRNA Editing Enhancement Activity from Chicken Enterocytes 1 Makoto Nakamuta,* An Tsai,† Lawrence Chan,‡ Nicholas O. Davidson,§ and Ba-Bie Teng† ,2 †Research Center for Human Genetics, Institute of Molecular Medicine, University of Texas—Houston, Houston, Texas 77030; *Department of Internal Medicine III, Kyushu University, Fukuoka, Japan; ‡Department of Cell Biology and Medicine, Baylor College of Medicine, One Baylor Plaza, Houston, Texas 77030; and §Department of Medicine, University of Chicago, Chicago, Illinois 60637
Received December 8, 1998
Mammalian intestinal apolipoprotein B (apoB) mRNA edits codon 2153 from CAA in apoB100 mRNA to a stop codon (UAA) in apoB48 mRNA. By contrast, chicken intestinal apoB mRNA contains a CAA codon at the corresponding site, but is not edited. Chicken enterocyte S100 extracts fail to edit mammalian apoB RNA, but contain factor(s) which enhance the mammalian enterocytes editing activity. By converting the chicken apoB mooring sequences to the conserved mammalian sequences, the study confirmed that this 11-nucleotide stretch was necessary and sufficient for minimal RNA editing. Using rat and chicken apoB chimeric constructs, the study revealed that mammalian apoB sequences were required for editing enhancement. In concert with the 29-nucleotide conserved cassette, the 5* rat apoB element (nucleotides 6615– 6629) increased editing at C-6666, and was necessary for editing enhancement of chicken enterocyte S100 extracts. Similarly, the 3* rat apoB element (nucleotides 6726 – 6752) was required for editing enhancement of chicken enterocyte S100 extracts, but to a lesser extent in efficiency, compared to the 5* region. In conclusion, this study identified the sequences required for editing enhancement activity from chicken enterocyte S100 extracts. © 1999 Academic Press
Apolipoprotein B (ApoB) mRNA editing changes a specific nucleotide, C-6666 to a U, and is mediated by a multicomponent enzyme complex with the catalytic component, called APOBEC1, [1]. APOBEC1 is a cyti1
This is publication number 140-IMM from the Institute of Molecular Medicine for the Prevention of Human Diseases, University of Texas-Houston Health Science Center. 2 Corresponding author. Fax: 713-500-2424. E-mail: bteng@imm2. imm.uth.tmc.edu. Abbreviations used: apoB, apolipoprotein B; PAGE, polyacrylamide gel electrophoresis; kDa, kilodalton; SDS, sodium dodecyl sulfate; nt., nucleotide(s); PCR, polymerase chain reaction; 0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
dine deaminase-like enzyme that requires complementation factor(s) for catalytic activity [1–9]. Driscoll et al. [10] developed a simple, in vitro editing assay which allows the mapping of the sequence elements required for editing in the vicinity of the edited base C-6666. There is a 29-nucleotide cassette (nucleotides 66626690) flanking C-6666 which is fully conserved among mammals [11]. Immediately 59 of the edited base C (nucleotides 6661-6665) is a regulator region, immediately 39 of C-6666 is a spacer sequence (nucleotides 6667-6670), and further downstream of C-6666 (nucleotides 6671-6681) is a relatively invariant sequence, called a mooring sequence, that is essential for editing in vitro (reviews, [12, 13]). This 11-nucleotides mooring sequence is both necessary and sufficient for the editing of a C located 3-5 bases upstream, providing the flanking sequence contains a sufficiently AU-rich bulk RNA sequences [14 –16]. In addition, a region further upstream to C-6666 (nucleotides 6648-6661) has been found to modulate the efficiency of editing [17]. Recent work by Hersberger and Innerarity [18] has shown that in combination with the mooring sequence, 59 regions of nucleotides 6609-6628 and 6629-6640, and 39 region of nucleotides 6717-6747 increased editing of C-6666, with the upstream sequence of C-6666 being more efficient than the downstream sequence. We have previously studied chicken apoB gene expression to obtain insight into the evolution of apoB mRNA editing. We showed that chicken apoB contained a CAA codon at a position equivalent to mammalian codon 2153, but chicken apoB mRNA is not edited [5]. Compared to the mammalian mooring sequence, there are three nucleotide substitutions in the chicken sequences. Interestingly, the study revealed that chicken enterocyte S100 extracts enhanced the in vitro editing activity of mammalian enterocyte S100 extracts. This phenomenon was termed editing enhancement activity. The editing enhancement activity from chicken enterocyte S100 extracts is tissue-
744
Vol. 254, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS In vitro apoB RNA editing assay. The in vitro editing assay was carried out as described by Teng and Davidson [5], using 2 ng (or indicated amount) of a 160 nucleotide synthetic RNA template (nts. 6597-6756) in the presence of indicated amount of rat enterocyte S100 extract. The products of primer extension were fractionated on an 8% polyacrylamide-urea gel (National Diagnostics, GA) and radiolabeled bands of edited and unedited transcripts were quantitated either by scanning laser densitometry of the autoradiograph, or by PhosphorImager SF (Molecular Dynamics, CA), as indicated in the Figures.
specific, heat-sensitive, substrate-saturable, and sensitive to proteinase K, but it is resistant to micrococcal nuclease. The factor(s) responsible for this activity was partially purified and was found to have an average molecular mass of 49 kDa [5]. The biological implications of finding apoB mRNA editing enhancement activity from chicken enterocytes in which endogeneous apoB mRNA is not edited reflects a functional counterpart to the evolutionary emergence of the apoB mRNA editing machinery. By using the chicken apoB mRNA sequence, we took the opportunity to test hypotheses concerning critical regions of sequence diversity in an evolutionary context. In this study we analyzed the sequence requirements for editing enhancement activity. The data presented here show that either mooring sequence alone, or the 29-nucleotide conserved cassette, do not confer editing enhancement. Editing enhancement requires mammalian apoB mRNA sequences flanking the conserved region. Our findings demonstrate that sequences either upstream (nucleotides 6615-6629) or downstream (nucleotides 67266752) of nucleotide C-6666 in mammalian apoB RNA contain sufficient information to induce editing enhancement activity.
MS1A MS1G MS1C MS2AG MS2AC MS2GC MS3 R1 C1 RC CR
MATERIALS AND METHODS
RESULTS
Plasmid construction. A 160-base pair fragment of wild-type chicken apoB cDNA (nt. 6597-6756) was cloned into a pAlter vector (Promega, WI). By using Altered Sites In Vitro Mutagenesis System (Promega, WI), we made single, double, and triple nucleotides substitutions of the mooring sequences as shown in Fig. 1A (single nucleotide substitution, pMS1A 5 nt. 6671 A 3 T, pMS1G 5 nt. 6674 G 3 T, pMS1C 5 nt. 6680 C 3 T; double nucleotides substitution, pMS2AG 5 nts. 6671 A 3 T and 6674 G 3 T, pMS2GC 5 nts. 6674 G 3 T and 6680 C 3 T, pMS2AC 5 nts. 6671 A 3 T and 6680 C 3 T; and triple nucleotides substitution, pMS3 5 nts. 6671 A 3 T, 6674 G 3 T, and 6680 C 3 T). We also made a mutant construct, pMB, which replaced the 31-nucleotide bases (nt. 6660-6690) of chicken apoB sequences with the corresponding rat apoB sequences. In here, the next upstream C beyond C-6666 of pMB is at nt. 6648. The chimeric constructs of pRC and pCR were engineered using gene overlapping extension PCR methods [19] by using oligonucleotides R1, C1, and RC or CR with pfu DNA polymerase (Stratgene, CA). pRC contains rat apoB nucleotides 6597 to 6661 upstream of mammalian 29-nucleotide conserved cassette (MC-29, nts. 66626690) and chicken apoB nucleotides 6691 to 6756 downstream of the 29-nt cassette. In this construct, the next upstream C beyond C-6666 of pRC is at nt. 6661. pCR contains chicken apoB nucleotides 6597 to 6659 and rat apoB nucleotides 6691 to 6756 flanking the conserved sequences of pMB (nts. 6660-6690). In here, the next upstream C beyond C-6666 of pCR is at nt. 6648. In vitro RNA synthesis was carried out using each construct as a template.
Conversion of Chicken apoB mRNA Sequences to Mammalian apoB mRNA
Deletion plasmid construct. Chimeric construct pRC was digested with SmaI and SacI, whereas chimeric construct pCR was digested with XbaI and AatII. Deletion constructs were created by using Erase-a-Base System (Promega, WI) according to the manufacturer’s procedures and confirmed by sequencing. Preparation of enterocyte S100 extracts. Rat or chicken enterocyte cytosolic S100 extracts were prepared as described by Teng and Davidson [5]. These extracts were stored at 280°C and remained stable for many years.
In vitro apoB RNA editing enhancement assay. This assay was performed in exactly the same manner as the in vitro editing assay described above, except that an indicated amount of chicken enterocyte cytosolic S100 extract (5 mg) was added to each assay [5]. Oligonucleotides. Oligonucleotides used in this study were listed as follow. The lower case represented the mutated nucleotide: 59 59 59 59 59 59 59 59 59 59 59
GGTGCAAATtGAGCAGTAC GCAAATAGAtCAGTACATC AGAGCAGTAtATCAAAGAG TGGTGCAAATtGAtCAGTACATCA TTGAGCAGTAtATCAAAGAG TGCAAATAGAtCAGTATATC TTGATCAGTAtATCAAAGAG CTAATTGCCTTAGATAGTGC TCATTTTTTCAACTATTTTG TCTCAACTTGAGACATACGCGATACAATTTGATCAGTATA CTCCAGCTTCAAGTGTATCTGATACAATTTGATCAGTATA
We have previously demonstrated that the addition of chicken enterocyte S100 extract to rat enterocyte S100 extract increased the editing of rat apoB mRNA to 4-fold, compared to rat enterocyte S100 extract alone [5]. This editing enhancement activity from chicken enterocyte S100 extracts was not observed by using chicken apoB mRNA as template. To identify the important sequences required for apoB mRNA editing enhancement activity, we took an approach mimic the evolution by conversion chicken apoB mRNA sequences to mammalian apoB mRNA. Chicken apoB RNA has three nucleotides that are different from mammalian species in the mooring sequence region (Fig. 1A). We engineered single nucleotide, double nucleotides, and triple nucleotides substitutions to modify chicken nucleotides to mammalian sequences. Rat enterocyte S100 extracts alone or in addition of chicken enterocyte S100 extracts did not edit mutant RNAs from either single nucleotide or double nucleotides substitutions. As shown in Fig. 1B and Table 1, rat enterocyte S100 extract alone edited 8.7 6 1.3% of triple nucleotides substitution MS3 mutant RNA, whereas the addition of chicken enterocyte S100 extract decreased the edited MS3-RNA to 3.8 6 0.4%. When we used MB mutant RNA as template, the editing and editing enhancement activities were similar (Fig. 1B, and Table 1, 10.0 6 1.6%, 12.0 6 1.3%, respectively).
745
Vol. 254, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Mammalian apoB Sequence Is Required for In Vitro Editing Enhancement Activity
FIG. 1. In vitro editing and editing enhancement assays of mutant chicken apoB RNAs. A). Aligned 29-nucleotide cassette of rat and chicken apoB sequences (nucleotides 6660-6690). The editing codon CAA is highlighted and the 11-nucleotide mooring sequence (MS) is underlined. Hyphens represent nucleotides identical to the rat sequence. The three nucleotides that differ from chicken to rat in the mooring sequence region are indicated (*). B). In vitro editing (5 mg rat enterocyte S100 extract only), or editing enhancement (5 mg rat plus 5 mg chicken enterocyte S100 extracts) assays were performed with substrates of MS3-RNA (2 ng), triple nucleotides substitution of chicken apoB RNA at nts 6671 A 3 T, 6674 G 3 T, and 6680 C 3 T, or MB-RNA (2 ng), nts. 6662-6690 of chicken apoB sequences was replaced with the corresponding rat apoB sequences and in addition, the chicken apoB nucleotide C-6660 was substituted with an A. The reaction was incubated with buffer only (2), or 5 mg of rat enterocyte S100 extract (1), or in the presence of 5 mg of rat (1) plus chicken (1) enterocyte S100 extracts. The edited and unedited apoB RNAs were determined by primer extension analysis. The extent of editing was quantitated by scanning laser densitometry of autoradigram. The edited RNA (UAAc-ms3 and UAAc-mb), the unedited RNA (CAA), and primers are indicated.
Therefore, by converting chicken nucleotides 6671 A 3 T, 6674 G 3 T, and 6680 C 3 T simultaneously to the mammalian mooring sequences, chicken apoB RNA is edited. This data supports the conclusion that the mooring sequence is necessary and sufficient for RNA editing. However, the mooring sequence, or the 29nucleotide mammalian conserved cassette alone (MB) confers only minimal editing; it does not have enough information for editing enhancement activity. Therefore, other mammalian apoB sequences are required to enhance in vitro RNA editing activity in the presence of chicken enterocyte S100 extracts.
To investigate the sequence requirement for editing enhancement activity, we engineered chimeric apoB constructs by flanking the 29 nucleotide conserved region (MC-29) with rat (R) and chicken (C) apoB sequences. Construct pRC contained the 29 nucleotide conserved sequence (MC-29, 6662-6690) which was flanked by rat apoB nucleotides 6597-6661 in the 59 end and chicken apoB nucleotides 6691-6756 in the 39 end (Fig. 2A). Construct pCR contained the conserved sequence of nucleotides 6662-6690 (MC-29) plus the chicken nt. C-6660 was substituted to an A. This stretch of sequences was flanked by chicken apoB nucleotides 6597-6659 in the 59 end and rat apoB nucleotides 6691-6756 in the 39 end (Fig. 2B). In vitro RNA editing (rat enterocyte S100 extracts alone) and editing enhancement (rat plus chicken enterocytes S100 extracts) assays were carried out using 2 (0.015 pmole), 5, and 10 ng of synthetic RNA from pRC or pCR. A representative data using RC-RNA as substrate is shown in Fig. 2A and the average of five experiments are summarized in Table 1. The addition of chicken enterocyte S100 extract increased the edited RC-RNA 2-3 folds. In this experiment, increased of RNA template resulted in progressive reduction in both editing and editing enhancement efficiency, suggests a critical stoichiometry between complementation factor(s) of chicken enterocyte S100 extracts, rat-APOBEC1, and RNA-template for optimum editing efficiency. Similarly, in the presence of 2, 5, and 10 ng CR-RNA, the addition of chicken enterocyte S100 extracts also increased the edited CR-RNA to 2-3 folds (Fig. 2B and Table 1). Therefore, the data demonstrated that pro-
TABLE 1
Summary of Assay Results Editing
Editing Enhancement
% UAA (mean 6 s.d.)
Substrate RNA
n
MS3-RNA MB-RNA RC-RNA: 2 ng 5 ng 10 ng CR-RNA: 2 ng 5 ng 10 ng
5 5
8.7 6 1.3 10.0 6 1.6
3.8 6 0.4 12.0 6 1.3
5 5 5
29.5 6 1.7 21.3 6 5.5 18.5 6 8.5
73.1 6 6.7 46.3 6 2.2 38.7 6 4.9
5 5 5
12.7 6 5.1 19.2 6 2.6 18.1 6 2.3
34.1 6 3.1 28.4 6 4.0 29.2 6 6.7
Note. The results of in vitro apoB RNA editing and editing enhancement activities are shown for MS3-RNA, MB-RNA, RC-RNA, and CR-RNA. In vitro RNA editing or editing activity is presented quantitatively as % UAA. The data are shown as an average of five experiments 6 standard deviation.
746
Vol. 254, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
efficient when rat apoB sequences are placed in the 59 end of MC-29. Minimal Sequence Requirement for Efficient in Vitro Editing and Editing Enhancement Activities
FIG. 2. In vitro editing and editing enhancement assays with chimeric rat apoB and chicken apoB RNAs. A). Indicated amounts (2, 5, and 10 ng) of chimeric 59 rat apoB-MC29-chicken apoB RNA (pRC) were used as substrate to perform editing (5 mg rat enterocyte S100 extract only) and editing enhancement (5 mg of rat plus 5 mg of chicken enterocyte S100 extracts) assays. The edited RNA (UAAr) and unedited RNA (CAA) were determined by primer extension analysis. B). Indicated amounts (2, 5, and 10 ng) of chimeric 59 chicken apoB-MC29-rat apoB RNA (pCR) were used as substrate to perform editing (5 mg rat enterocyte S100 extract only) and editing enhancement (5 mg of rat plus 5 mg of chicken enterocyte S100 extracts) assays. The edited RNA (UAAc) and unedited RNA (CAA) were determined by primer extension analysis. The extent of editing was determined by scanning laser densitometry of autoradigram. The edited RNAs (UAAr and UAAc), the unedited RNA, and the primer are indicated.
viding rat apoB sequences at either 59 or 39 ends of MC-29 conserved cassette increased the proportion of RNA editing. Editing enhancement appears to be more
To determine the minimal rat apoB flanking sequences required for efficient apoB RNA editing and editing enhancement activity, we deleted rat apoB sequences from the 59 or the 39 ends that flanking the MC-29 cassette as described in Methods. As shown in Fig. 3, rat apoB sequences from the 59-end of pRC were deleted from nucleotide 6597 to nucleotides 6603, 6616, 6630, 6636, and 6658, denoted as pRC1, pRC2, pRC3, pRC4, and pRC5, respectively. Rat apoB sequences from the 39-end of pCR were deleted from nucleotide 6756 to nucleotides 6752, 6726, 6718, and 6705, denoted as pCR1, pCR2, pCR3, and pCR4, respectively. To confirm whether the RNA length in this study will influence RNA editing or editing enhancement, we used rat apoB RNA varied from 280 to 100 nucleotides in length as substrate for the assays. In the control experiment the data demonstrated that there were no significant difference in the amounts of editing or editing enhancement from the substrates we used (data not shown). Equal amounts of synthetic RNA (0.015 pmole) from pRC or pCR series were used and the results are shown in Fig. 3. Compared to parental RC-RNA, rat enterocyte extracts (editing activity) edited similar amounts of deletion RNAs of RC1 and RC2 (27.7 6 3.3% and 25.8 6 1.4%, respectively), and the addition of chicken enterocyte extracts (editing enhancement activity) increased the proportion of editing to 77.16 6.9% and 67.3 6 6.3%, respectively. When the construct was shortened to position 6629 (pRC3), rat enterocyte S100 extracts edited only 8.5 6 2.9% of RC3-RNA and the addition of chicken enterocyte S100 extracts increased the proportion of edited RC3-RNA modestly to 14.9 6 3.3%. Further deletion of rat apoB nucleotides to position 6635 and 6657, pRC4 and pRC5, respectively, did not result in further decrease of edited RNA, compared to that of the RC3-construct. However, editing enhancement in these constructs was eliminated. In conclusion, the data suggest that nucleotides 6615-6629 of rat apoB are important for both efficient editing and editing enhancement activities. To further examine the role of flanking sequence requirements, a parallel set of experiments was undertaken with the pCR constructs. Rat enterocyte S100 extracts edited 18.4 6 6.2% of CR1-RNA and the addition of chicken enterocyte S100 extracts enhanced RNA editing to 34.6 6 1.5%. Both editing and editing enhancement activities of pCR1-RNA were similar to that of the parental pCR-RNA (15.7 6 5.1% and 32.0 6 1.1%, respectively). Following deletion of the RNA to nucleotide 6726 (CR2-RNA), rat enterocyte S100 ex-
747
Vol. 254, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. Effects of chicken and rat chimeric RNAs on in vitro editing and editing enhancement activities. The data are expressed as percentage of edited RNA (mean 6 s.d., n 5 5). Equal molar (0.015 pmole) of deleted chimeric RNAs, RC-RNA or CR-RNA, were used as substrate for in vitro editing and editing enhancement assays. Construct pRC contains a 29-nt conserved cassette (nts. 6662-6690, MC-29) which is flanked by rat apoB sequence (6597-6661) in the 59 end and chicken apoB sequence (6691-6756) in the 39 end. pRC was deleted from 59 to nucleotides 6603, 6615, 6629, 6635, and 6657, termed pRC1, pRC2, pRC3, pRC4, and pRC5, respectively. Construct pCR contains a 29-nt conserved cassette (nts. 6662-6690, MC-29) plus chicken apoB nucleotide C-6660 was substituted with an A. This region was flanked by 59 chicken apoB sequence (6597-6659) and 39 rat apoB sequence (6691-6756). pCR was deleted from 39 to nucleotides 6752, 6726, 6718, and 6705, termed pCR1, pCR2, pCR3, and pCR4, respectively. The lengths of apoB RNA and polylinker of each construct are shown.
tracts only edited 7.6 6 1.7% of CR2-RNA, and the addition of chicken enterocyte extracts increased the edited CR2-RNA slightly to 10.8 6 2.5%. Upon further shortening of the RNA length to nucleotides 6718, and 6705, CR3-RNA and CR4-RNA, respectively, the editing activity remained at ;7% and editing enhancement activity at ;10%. Taken together, the data suggest that rat apoB RNA sequences (nucleotides 67266756) 39 to the edited base (C-6666) are needed for efficient editing and editing enhancement activities at the canonical site. DISCUSSION The 11-nt mooring sequence (nt. 6671-6681) has been previously characterized [14, 15, 20]. Mutations in this region drastically reduce editing by rat enterocyte S100 extracts [14, 15, 20] and by baboon enterocyte S100 extracts [17]. Driscoll et al. [17] showed that the mooring sequence can induce editing of other cytidines in apoB mRNA, as well as in a heterologous luciferase mRNA. In agreement with these observa-
tions, we showed that by converting the nucleotides from chicken mooring sequence to mammalian mooring sequence, chicken apoB RNA was edited (;8.7% UAA). However, using the same approach, Anant et al. [21] have shown that by converting the chicken mooring sequence to rat mooring sequence there was no editing at all. They observed ;4% editing activity only when the entire 29-nucleotide cassette of chicken apoB was converted to mammalian conserved sequences (like pMB). The difference in the results of our study and that of Anata et al., may be due to the differences in the materials we used. Anant et al. [21] used GSTfusion APOBEC1 plus chicken enterocyte S100 extracts rather than rat enterocyte S100 extracts used in this study. The GST-fusion APOBEC1 has been shown to have very low editing activity. Unlike other investigators who have used mutations of mammalian apoB sequences to demonstrate the significance of mooring sequence in RNA editing [14, 20], we converted chicken apoB sequences to rat apoB sequences and our data confirmed the importance and necessity of mooring sequences in RNA editing. The nucleotide changes in
748
Vol. 254, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Aligned nucleotide sequences of the apoB mRNA editing regions in five species. The numbering corresponds to the published human apoB100 sequence [23]. The chicken sequence is provided on the top line, and nucleotide differences from the chicken sequences in the other species compared are aligned beneath. Hyphens represent nucleotides identical to the chicken sequence. The nucleotides corresponding to apoB mRNA editing codon and mooring sequence are underlined. The 29-nucleotide conserved cassette (MC-29) is marked with a square. The efficiency elements (nts. 6609-6640 and 6717-6747) described by Hersberger and Innerarity [18] are marked with a square. Elements required for editing and editing enhancement (nts. 6615-6628 and 6726-6752) are highlighted with the shade.
this region during evolution have profound consequences on RNA editing. Our study demonstrates that neither the 11-nt mooring sequence, nor the entire 29-nt conserved cassette alone were sufficient for editing enhancement activity in chicken enterocyte S100 extracts. In order to elucidate the sequence element(s) required for in vitro editing enhancement activity, we engineered chimeric rat apoB/chicken apoB (pRC), or chicken apoB/rat apoB (pCR) constructs. Using these chimeric RNAs, rat enterocyte S100 extracts edited RC-RNA twice as efficiently as CR-RNA (28% vs. 15%), and in the presence of chicken enterocyte extracts, the proportion of edited RC-RNA, and CR-RNA were increased to ;77% and ;35%, respectively. Therefore, replacement of rat apoB RNA on either the 59 or 39 end of the 29-nt conserved cassette provides the sequence elements required for chicken editing enhancement activity. Using deletion strategy on pRC, we showed that nucleotides 6615-6629 were important for inducing editing and editing enhancement activity. Like the mooring sequence or the 29-nt conserved cassette, this region (nts. 6615-6628) is A/T rich (A 1 T 5 62%) and fully conserved across mammalian species (Fig. 4). Comparison of this stretch (nts. 6615-6628) to chicken sequences reveals 5 nucleotide substitutions. The significance of this region was also demonstrated by Hersberger and Innerarity [18], who showed that rabbit apoB nucleotides 6609-6629 contains important elements for editing at C-6666. By replacing rat apoB sequence (nucleotides 66906756) in the 39 of the 29-nt conserved cassette, chicken enterocyte S100 extracts also enhanced the proportion of editing of chimeric CR-RNA. Deletion studies revealed that nucleotides 6726-6752 provided sufficient information for apoB RNA editing and editing enhancement. Unlike the 59 upstream element, this region is much more diverged among
different vertebrate species, but it is A/T rich (A 1 T 5 73%) (Fig. 4). This may in part account for the finding that chicken enterocyte S100 extracts enhance rat, pig, and human RNAs with various editing efficiencies in vitro [5]. Interestingly, Hersberger and Innerarity also located a 31-nt fragment (nucleotides 6717-6747) which increased apoB RNA editing of C-6666 [18]. Sowden et al have proposed a “population gating” hypothesis to explain the apoB RNA editing mechanism by which cis- and transacting factor interact to determine editing efficiency [22]. The “gate” (edtosomes) evaluates each RNA passing through. Presumably, 59 element of nts. 6615-6628 and 39 element of nts. 6726-6752 provide the characteristics for increasing the efficiency of RNA editing and editing enhancement of C-6666. ACKNOWLEDGMENTS This research was supported by National Institute of Health Grants HL-53441 (to BBT) and HL-56668 (to LC) and American Heart Association Grant-In-Aid (to BBT). We are grateful to Dr. Kunihisa Kobayashi from Baylor College of Medicine for the preparation of computer graphics. We are also grateful to Dr. Eva Zsigmond from Institute of Molecular Medicine, University of TexasHouston for critical reading this manuscript.
REFERENCES
749
1. Teng, B. B., Burant, C. F., and Davidson, N. O. (1993) Science 260, 1816 –1819. 2. Navaratnam, N., Morrison, J. R., Bhattacharya, S., Patel, D., Funahashi, T., Giannon, F., Teng, B. B., Davidson, N. O., and Scott, J. (1993) J. Biol. Chem. 268, 20709 –20712. 3. Giannoni, F., Chou, S.-C., Skarosi, S. F., Verp, M. S., Field, F. J., Coleman, R. A., and Davidson, N. O. (1995) J. Lipid Res. 36, 1664 –1675. 4. Driscoll, D. M., and Zhang, Q. (1994) J. Biol. Chem. 269, 19843– 19847.
Vol. 254, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
5. Teng, B. B., and Davidson, N. O. (1992) J. Biol. Chem. 267, 21265–21272. 6. Yamanaka, S., Poksay, K. S., Balestra, M. E., Zeng, G.-Q., and Innerarity, T. L. (1994) J. Biol. Chem. 269, 21725–21734. 7. Schock, D., Kuo, S.-R., Steinburg, M. F., Bolognino, M., Sparks, J. D., Sparks, C., and Smith, H. C. (1996) Proc. Natl. Acad. Sci. USA 93, 1097–1102. 8. Yang, Y., Yang, Y., Kovalski, K., and Smith, H. C. (1997) J. Biol. Chem. 272, 27700 –27706. 9. Lau, P. P., Zhu, H.-J., Nakamuta, M., and Chan, L. (1997) J. Biol. Chem. 272, 1452–1455. 10. Driscoll, D. M., Wynne, J. K., Wallis, S. C., and Scott, J. (1989) Cell 58, 519 –525. 11. Teng, B. B., Black, D. D., and Davidson, N. O. (1990) Biochem. Biophys. Res. Commun. 173, 74 – 80. 12. Hodges, P., and Scott, J. (1992) Trends Biochem. Sci. 17, 77– 81. 13. Smith, H. C. (1993) Seminars in Cell Biol. 4, 267–278. 14. Backus, J. W., and Smith, H. C. (1991) Nuc. Acids Res. 19, 6781– 6786.
15. Backus, J. W., and Smith, H. (1992) Nuc. Acids Res. 20, 6007– 6014. 16. Backus, J. W., Schock, D., and Smith, H. C. (1994) Biochim. Biophys. Acta 1219, 1–14. 17. Driscoll, D. M., Lakhe-Reedy, S., Oleksa, L. M., and Martinez, D. (1993) Mol. Cellular Biol. 13, 7288 –7294. 18. Hersberger, M., and Innerarity, T. L. (1998) J. Biol. Chem. 273, 9435–9442. 19. Stevenson, S. C., Marshall-Neff, J., Teng, B. B., Lee, C. B., Roy, S., and McClelland, A. (1995) Arteriosclerosis & Thrombosis 15, 479 – 484. 20. Shah, R. R., Knott, T. J., Legros, J. E., Navaratnam, N., Greeve, J. C., and Scott, J. (1991) J. Biol. Chem. 266, 16301–16304. 21. Anant, S., MacGinnitie, A. J., and Davidson, N. O. (1995) J. Biol. Chem. 270, 14762–14767. 22. Sowden, M., Hamm, J. K., Spinelli, S., and Smith, H. C. (1996) RNA 2, 274 –288. 23. Knott, T. J., Wallis, S. C., Powell, L. M., Pease, R. J., Lusis, A. J., Blackhart, B., McCarthy, B. J., Mahley, R. W., Levy-Wilson, B., and Scott, J. (1986) Nucleic Acids Res. 14, 7501–7503.
750
Sequence Elements Required for Apolipoprotein B mRNA Editing Enhancement Activity from Chicken Enterocytes 1 Makoto Nakamuta,* An Tsai,† Lawrence Chan,‡ Nicholas O. Davidson,§ and Ba-Bie Teng† ,2 †Research Center for Human Genetics, Institute of Molecular Medicine, University of Texas—Houston, Houston, Texas 77030; *Department of Internal Medicine III, Kyushu University, Fukuoka, Japan; ‡Department of Cell Biology and Medicine, Baylor College of Medicine, One Baylor Plaza, Houston, Texas 77030; and §Department of Medicine, University of Chicago, Chicago, Illinois 60637
Received December 8, 1998
Mammalian intestinal apolipoprotein B (apoB) mRNA edits codon 2153 from CAA in apoB100 mRNA to a stop codon (UAA) in apoB48 mRNA. By contrast, chicken intestinal apoB mRNA contains a CAA codon at the corresponding site, but is not edited. Chicken enterocyte S100 extracts fail to edit mammalian apoB RNA, but contain factor(s) which enhance the mammalian enterocytes editing activity. By converting the chicken apoB mooring sequences to the conserved mammalian sequences, the study confirmed that this 11-nucleotide stretch was necessary and sufficient for minimal RNA editing. Using rat and chicken apoB chimeric constructs, the study revealed that mammalian apoB sequences were required for editing enhancement. In concert with the 29-nucleotide conserved cassette, the 5* rat apoB element (nucleotides 6615– 6629) increased editing at C-6666, and was necessary for editing enhancement of chicken enterocyte S100 extracts. Similarly, the 3* rat apoB element (nucleotides 6726 – 6752) was required for editing enhancement of chicken enterocyte S100 extracts, but to a lesser extent in efficiency, compared to the 5* region. In conclusion, this study identified the sequences required for editing enhancement activity from chicken enterocyte S100 extracts. © 1999 Academic Press
Apolipoprotein B (ApoB) mRNA editing changes a specific nucleotide, C-6666 to a U, and is mediated by a multicomponent enzyme complex with the catalytic component, called APOBEC1, [1]. APOBEC1 is a cyti1
This is publication number 140-IMM from the Institute of Molecular Medicine for the Prevention of Human Diseases, University of Texas-Houston Health Science Center. 2 Corresponding author. Fax: 713-500-2424. E-mail: bteng@imm2. imm.uth.tmc.edu. Abbreviations used: apoB, apolipoprotein B; PAGE, polyacrylamide gel electrophoresis; kDa, kilodalton; SDS, sodium dodecyl sulfate; nt., nucleotide(s); PCR, polymerase chain reaction; 0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
dine deaminase-like enzyme that requires complementation factor(s) for catalytic activity [1–9]. Driscoll et al. [10] developed a simple, in vitro editing assay which allows the mapping of the sequence elements required for editing in the vicinity of the edited base C-6666. There is a 29-nucleotide cassette (nucleotides 66626690) flanking C-6666 which is fully conserved among mammals [11]. Immediately 59 of the edited base C (nucleotides 6661-6665) is a regulator region, immediately 39 of C-6666 is a spacer sequence (nucleotides 6667-6670), and further downstream of C-6666 (nucleotides 6671-6681) is a relatively invariant sequence, called a mooring sequence, that is essential for editing in vitro (reviews, [12, 13]). This 11-nucleotides mooring sequence is both necessary and sufficient for the editing of a C located 3-5 bases upstream, providing the flanking sequence contains a sufficiently AU-rich bulk RNA sequences [14 –16]. In addition, a region further upstream to C-6666 (nucleotides 6648-6661) has been found to modulate the efficiency of editing [17]. Recent work by Hersberger and Innerarity [18] has shown that in combination with the mooring sequence, 59 regions of nucleotides 6609-6628 and 6629-6640, and 39 region of nucleotides 6717-6747 increased editing of C-6666, with the upstream sequence of C-6666 being more efficient than the downstream sequence. We have previously studied chicken apoB gene expression to obtain insight into the evolution of apoB mRNA editing. We showed that chicken apoB contained a CAA codon at a position equivalent to mammalian codon 2153, but chicken apoB mRNA is not edited [5]. Compared to the mammalian mooring sequence, there are three nucleotide substitutions in the chicken sequences. Interestingly, the study revealed that chicken enterocyte S100 extracts enhanced the in vitro editing activity of mammalian enterocyte S100 extracts. This phenomenon was termed editing enhancement activity. The editing enhancement activity from chicken enterocyte S100 extracts is tissue-
744
Vol. 254, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS In vitro apoB RNA editing assay. The in vitro editing assay was carried out as described by Teng and Davidson [5], using 2 ng (or indicated amount) of a 160 nucleotide synthetic RNA template (nts. 6597-6756) in the presence of indicated amount of rat enterocyte S100 extract. The products of primer extension were fractionated on an 8% polyacrylamide-urea gel (National Diagnostics, GA) and radiolabeled bands of edited and unedited transcripts were quantitated either by scanning laser densitometry of the autoradiograph, or by PhosphorImager SF (Molecular Dynamics, CA), as indicated in the Figures.
specific, heat-sensitive, substrate-saturable, and sensitive to proteinase K, but it is resistant to micrococcal nuclease. The factor(s) responsible for this activity was partially purified and was found to have an average molecular mass of 49 kDa [5]. The biological implications of finding apoB mRNA editing enhancement activity from chicken enterocytes in which endogeneous apoB mRNA is not edited reflects a functional counterpart to the evolutionary emergence of the apoB mRNA editing machinery. By using the chicken apoB mRNA sequence, we took the opportunity to test hypotheses concerning critical regions of sequence diversity in an evolutionary context. In this study we analyzed the sequence requirements for editing enhancement activity. The data presented here show that either mooring sequence alone, or the 29-nucleotide conserved cassette, do not confer editing enhancement. Editing enhancement requires mammalian apoB mRNA sequences flanking the conserved region. Our findings demonstrate that sequences either upstream (nucleotides 6615-6629) or downstream (nucleotides 67266752) of nucleotide C-6666 in mammalian apoB RNA contain sufficient information to induce editing enhancement activity.
MS1A MS1G MS1C MS2AG MS2AC MS2GC MS3 R1 C1 RC CR
MATERIALS AND METHODS
RESULTS
Plasmid construction. A 160-base pair fragment of wild-type chicken apoB cDNA (nt. 6597-6756) was cloned into a pAlter vector (Promega, WI). By using Altered Sites In Vitro Mutagenesis System (Promega, WI), we made single, double, and triple nucleotides substitutions of the mooring sequences as shown in Fig. 1A (single nucleotide substitution, pMS1A 5 nt. 6671 A 3 T, pMS1G 5 nt. 6674 G 3 T, pMS1C 5 nt. 6680 C 3 T; double nucleotides substitution, pMS2AG 5 nts. 6671 A 3 T and 6674 G 3 T, pMS2GC 5 nts. 6674 G 3 T and 6680 C 3 T, pMS2AC 5 nts. 6671 A 3 T and 6680 C 3 T; and triple nucleotides substitution, pMS3 5 nts. 6671 A 3 T, 6674 G 3 T, and 6680 C 3 T). We also made a mutant construct, pMB, which replaced the 31-nucleotide bases (nt. 6660-6690) of chicken apoB sequences with the corresponding rat apoB sequences. In here, the next upstream C beyond C-6666 of pMB is at nt. 6648. The chimeric constructs of pRC and pCR were engineered using gene overlapping extension PCR methods [19] by using oligonucleotides R1, C1, and RC or CR with pfu DNA polymerase (Stratgene, CA). pRC contains rat apoB nucleotides 6597 to 6661 upstream of mammalian 29-nucleotide conserved cassette (MC-29, nts. 66626690) and chicken apoB nucleotides 6691 to 6756 downstream of the 29-nt cassette. In this construct, the next upstream C beyond C-6666 of pRC is at nt. 6661. pCR contains chicken apoB nucleotides 6597 to 6659 and rat apoB nucleotides 6691 to 6756 flanking the conserved sequences of pMB (nts. 6660-6690). In here, the next upstream C beyond C-6666 of pCR is at nt. 6648. In vitro RNA synthesis was carried out using each construct as a template.
Conversion of Chicken apoB mRNA Sequences to Mammalian apoB mRNA
Deletion plasmid construct. Chimeric construct pRC was digested with SmaI and SacI, whereas chimeric construct pCR was digested with XbaI and AatII. Deletion constructs were created by using Erase-a-Base System (Promega, WI) according to the manufacturer’s procedures and confirmed by sequencing. Preparation of enterocyte S100 extracts. Rat or chicken enterocyte cytosolic S100 extracts were prepared as described by Teng and Davidson [5]. These extracts were stored at 280°C and remained stable for many years.
In vitro apoB RNA editing enhancement assay. This assay was performed in exactly the same manner as the in vitro editing assay described above, except that an indicated amount of chicken enterocyte cytosolic S100 extract (5 mg) was added to each assay [5]. Oligonucleotides. Oligonucleotides used in this study were listed as follow. The lower case represented the mutated nucleotide: 59 59 59 59 59 59 59 59 59 59 59
GGTGCAAATtGAGCAGTAC GCAAATAGAtCAGTACATC AGAGCAGTAtATCAAAGAG TGGTGCAAATtGAtCAGTACATCA TTGAGCAGTAtATCAAAGAG TGCAAATAGAtCAGTATATC TTGATCAGTAtATCAAAGAG CTAATTGCCTTAGATAGTGC TCATTTTTTCAACTATTTTG TCTCAACTTGAGACATACGCGATACAATTTGATCAGTATA CTCCAGCTTCAAGTGTATCTGATACAATTTGATCAGTATA
We have previously demonstrated that the addition of chicken enterocyte S100 extract to rat enterocyte S100 extract increased the editing of rat apoB mRNA to 4-fold, compared to rat enterocyte S100 extract alone [5]. This editing enhancement activity from chicken enterocyte S100 extracts was not observed by using chicken apoB mRNA as template. To identify the important sequences required for apoB mRNA editing enhancement activity, we took an approach mimic the evolution by conversion chicken apoB mRNA sequences to mammalian apoB mRNA. Chicken apoB RNA has three nucleotides that are different from mammalian species in the mooring sequence region (Fig. 1A). We engineered single nucleotide, double nucleotides, and triple nucleotides substitutions to modify chicken nucleotides to mammalian sequences. Rat enterocyte S100 extracts alone or in addition of chicken enterocyte S100 extracts did not edit mutant RNAs from either single nucleotide or double nucleotides substitutions. As shown in Fig. 1B and Table 1, rat enterocyte S100 extract alone edited 8.7 6 1.3% of triple nucleotides substitution MS3 mutant RNA, whereas the addition of chicken enterocyte S100 extract decreased the edited MS3-RNA to 3.8 6 0.4%. When we used MB mutant RNA as template, the editing and editing enhancement activities were similar (Fig. 1B, and Table 1, 10.0 6 1.6%, 12.0 6 1.3%, respectively).
745
Vol. 254, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Mammalian apoB Sequence Is Required for In Vitro Editing Enhancement Activity
FIG. 1. In vitro editing and editing enhancement assays of mutant chicken apoB RNAs. A). Aligned 29-nucleotide cassette of rat and chicken apoB sequences (nucleotides 6660-6690). The editing codon CAA is highlighted and the 11-nucleotide mooring sequence (MS) is underlined. Hyphens represent nucleotides identical to the rat sequence. The three nucleotides that differ from chicken to rat in the mooring sequence region are indicated (*). B). In vitro editing (5 mg rat enterocyte S100 extract only), or editing enhancement (5 mg rat plus 5 mg chicken enterocyte S100 extracts) assays were performed with substrates of MS3-RNA (2 ng), triple nucleotides substitution of chicken apoB RNA at nts 6671 A 3 T, 6674 G 3 T, and 6680 C 3 T, or MB-RNA (2 ng), nts. 6662-6690 of chicken apoB sequences was replaced with the corresponding rat apoB sequences and in addition, the chicken apoB nucleotide C-6660 was substituted with an A. The reaction was incubated with buffer only (2), or 5 mg of rat enterocyte S100 extract (1), or in the presence of 5 mg of rat (1) plus chicken (1) enterocyte S100 extracts. The edited and unedited apoB RNAs were determined by primer extension analysis. The extent of editing was quantitated by scanning laser densitometry of autoradigram. The edited RNA (UAAc-ms3 and UAAc-mb), the unedited RNA (CAA), and primers are indicated.
Therefore, by converting chicken nucleotides 6671 A 3 T, 6674 G 3 T, and 6680 C 3 T simultaneously to the mammalian mooring sequences, chicken apoB RNA is edited. This data supports the conclusion that the mooring sequence is necessary and sufficient for RNA editing. However, the mooring sequence, or the 29nucleotide mammalian conserved cassette alone (MB) confers only minimal editing; it does not have enough information for editing enhancement activity. Therefore, other mammalian apoB sequences are required to enhance in vitro RNA editing activity in the presence of chicken enterocyte S100 extracts.
To investigate the sequence requirement for editing enhancement activity, we engineered chimeric apoB constructs by flanking the 29 nucleotide conserved region (MC-29) with rat (R) and chicken (C) apoB sequences. Construct pRC contained the 29 nucleotide conserved sequence (MC-29, 6662-6690) which was flanked by rat apoB nucleotides 6597-6661 in the 59 end and chicken apoB nucleotides 6691-6756 in the 39 end (Fig. 2A). Construct pCR contained the conserved sequence of nucleotides 6662-6690 (MC-29) plus the chicken nt. C-6660 was substituted to an A. This stretch of sequences was flanked by chicken apoB nucleotides 6597-6659 in the 59 end and rat apoB nucleotides 6691-6756 in the 39 end (Fig. 2B). In vitro RNA editing (rat enterocyte S100 extracts alone) and editing enhancement (rat plus chicken enterocytes S100 extracts) assays were carried out using 2 (0.015 pmole), 5, and 10 ng of synthetic RNA from pRC or pCR. A representative data using RC-RNA as substrate is shown in Fig. 2A and the average of five experiments are summarized in Table 1. The addition of chicken enterocyte S100 extract increased the edited RC-RNA 2-3 folds. In this experiment, increased of RNA template resulted in progressive reduction in both editing and editing enhancement efficiency, suggests a critical stoichiometry between complementation factor(s) of chicken enterocyte S100 extracts, rat-APOBEC1, and RNA-template for optimum editing efficiency. Similarly, in the presence of 2, 5, and 10 ng CR-RNA, the addition of chicken enterocyte S100 extracts also increased the edited CR-RNA to 2-3 folds (Fig. 2B and Table 1). Therefore, the data demonstrated that pro-
TABLE 1
Summary of Assay Results Editing
Editing Enhancement
% UAA (mean 6 s.d.)
Substrate RNA
n
MS3-RNA MB-RNA RC-RNA: 2 ng 5 ng 10 ng CR-RNA: 2 ng 5 ng 10 ng
5 5
8.7 6 1.3 10.0 6 1.6
3.8 6 0.4 12.0 6 1.3
5 5 5
29.5 6 1.7 21.3 6 5.5 18.5 6 8.5
73.1 6 6.7 46.3 6 2.2 38.7 6 4.9
5 5 5
12.7 6 5.1 19.2 6 2.6 18.1 6 2.3
34.1 6 3.1 28.4 6 4.0 29.2 6 6.7
Note. The results of in vitro apoB RNA editing and editing enhancement activities are shown for MS3-RNA, MB-RNA, RC-RNA, and CR-RNA. In vitro RNA editing or editing activity is presented quantitatively as % UAA. The data are shown as an average of five experiments 6 standard deviation.
746
Vol. 254, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
efficient when rat apoB sequences are placed in the 59 end of MC-29. Minimal Sequence Requirement for Efficient in Vitro Editing and Editing Enhancement Activities
FIG. 2. In vitro editing and editing enhancement assays with chimeric rat apoB and chicken apoB RNAs. A). Indicated amounts (2, 5, and 10 ng) of chimeric 59 rat apoB-MC29-chicken apoB RNA (pRC) were used as substrate to perform editing (5 mg rat enterocyte S100 extract only) and editing enhancement (5 mg of rat plus 5 mg of chicken enterocyte S100 extracts) assays. The edited RNA (UAAr) and unedited RNA (CAA) were determined by primer extension analysis. B). Indicated amounts (2, 5, and 10 ng) of chimeric 59 chicken apoB-MC29-rat apoB RNA (pCR) were used as substrate to perform editing (5 mg rat enterocyte S100 extract only) and editing enhancement (5 mg of rat plus 5 mg of chicken enterocyte S100 extracts) assays. The edited RNA (UAAc) and unedited RNA (CAA) were determined by primer extension analysis. The extent of editing was determined by scanning laser densitometry of autoradigram. The edited RNAs (UAAr and UAAc), the unedited RNA, and the primer are indicated.
viding rat apoB sequences at either 59 or 39 ends of MC-29 conserved cassette increased the proportion of RNA editing. Editing enhancement appears to be more
To determine the minimal rat apoB flanking sequences required for efficient apoB RNA editing and editing enhancement activity, we deleted rat apoB sequences from the 59 or the 39 ends that flanking the MC-29 cassette as described in Methods. As shown in Fig. 3, rat apoB sequences from the 59-end of pRC were deleted from nucleotide 6597 to nucleotides 6603, 6616, 6630, 6636, and 6658, denoted as pRC1, pRC2, pRC3, pRC4, and pRC5, respectively. Rat apoB sequences from the 39-end of pCR were deleted from nucleotide 6756 to nucleotides 6752, 6726, 6718, and 6705, denoted as pCR1, pCR2, pCR3, and pCR4, respectively. To confirm whether the RNA length in this study will influence RNA editing or editing enhancement, we used rat apoB RNA varied from 280 to 100 nucleotides in length as substrate for the assays. In the control experiment the data demonstrated that there were no significant difference in the amounts of editing or editing enhancement from the substrates we used (data not shown). Equal amounts of synthetic RNA (0.015 pmole) from pRC or pCR series were used and the results are shown in Fig. 3. Compared to parental RC-RNA, rat enterocyte extracts (editing activity) edited similar amounts of deletion RNAs of RC1 and RC2 (27.7 6 3.3% and 25.8 6 1.4%, respectively), and the addition of chicken enterocyte extracts (editing enhancement activity) increased the proportion of editing to 77.16 6.9% and 67.3 6 6.3%, respectively. When the construct was shortened to position 6629 (pRC3), rat enterocyte S100 extracts edited only 8.5 6 2.9% of RC3-RNA and the addition of chicken enterocyte S100 extracts increased the proportion of edited RC3-RNA modestly to 14.9 6 3.3%. Further deletion of rat apoB nucleotides to position 6635 and 6657, pRC4 and pRC5, respectively, did not result in further decrease of edited RNA, compared to that of the RC3-construct. However, editing enhancement in these constructs was eliminated. In conclusion, the data suggest that nucleotides 6615-6629 of rat apoB are important for both efficient editing and editing enhancement activities. To further examine the role of flanking sequence requirements, a parallel set of experiments was undertaken with the pCR constructs. Rat enterocyte S100 extracts edited 18.4 6 6.2% of CR1-RNA and the addition of chicken enterocyte S100 extracts enhanced RNA editing to 34.6 6 1.5%. Both editing and editing enhancement activities of pCR1-RNA were similar to that of the parental pCR-RNA (15.7 6 5.1% and 32.0 6 1.1%, respectively). Following deletion of the RNA to nucleotide 6726 (CR2-RNA), rat enterocyte S100 ex-
747
Vol. 254, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. Effects of chicken and rat chimeric RNAs on in vitro editing and editing enhancement activities. The data are expressed as percentage of edited RNA (mean 6 s.d., n 5 5). Equal molar (0.015 pmole) of deleted chimeric RNAs, RC-RNA or CR-RNA, were used as substrate for in vitro editing and editing enhancement assays. Construct pRC contains a 29-nt conserved cassette (nts. 6662-6690, MC-29) which is flanked by rat apoB sequence (6597-6661) in the 59 end and chicken apoB sequence (6691-6756) in the 39 end. pRC was deleted from 59 to nucleotides 6603, 6615, 6629, 6635, and 6657, termed pRC1, pRC2, pRC3, pRC4, and pRC5, respectively. Construct pCR contains a 29-nt conserved cassette (nts. 6662-6690, MC-29) plus chicken apoB nucleotide C-6660 was substituted with an A. This region was flanked by 59 chicken apoB sequence (6597-6659) and 39 rat apoB sequence (6691-6756). pCR was deleted from 39 to nucleotides 6752, 6726, 6718, and 6705, termed pCR1, pCR2, pCR3, and pCR4, respectively. The lengths of apoB RNA and polylinker of each construct are shown.
tracts only edited 7.6 6 1.7% of CR2-RNA, and the addition of chicken enterocyte extracts increased the edited CR2-RNA slightly to 10.8 6 2.5%. Upon further shortening of the RNA length to nucleotides 6718, and 6705, CR3-RNA and CR4-RNA, respectively, the editing activity remained at ;7% and editing enhancement activity at ;10%. Taken together, the data suggest that rat apoB RNA sequences (nucleotides 67266756) 39 to the edited base (C-6666) are needed for efficient editing and editing enhancement activities at the canonical site. DISCUSSION The 11-nt mooring sequence (nt. 6671-6681) has been previously characterized [14, 15, 20]. Mutations in this region drastically reduce editing by rat enterocyte S100 extracts [14, 15, 20] and by baboon enterocyte S100 extracts [17]. Driscoll et al. [17] showed that the mooring sequence can induce editing of other cytidines in apoB mRNA, as well as in a heterologous luciferase mRNA. In agreement with these observa-
tions, we showed that by converting the nucleotides from chicken mooring sequence to mammalian mooring sequence, chicken apoB RNA was edited (;8.7% UAA). However, using the same approach, Anant et al. [21] have shown that by converting the chicken mooring sequence to rat mooring sequence there was no editing at all. They observed ;4% editing activity only when the entire 29-nucleotide cassette of chicken apoB was converted to mammalian conserved sequences (like pMB). The difference in the results of our study and that of Anata et al., may be due to the differences in the materials we used. Anant et al. [21] used GSTfusion APOBEC1 plus chicken enterocyte S100 extracts rather than rat enterocyte S100 extracts used in this study. The GST-fusion APOBEC1 has been shown to have very low editing activity. Unlike other investigators who have used mutations of mammalian apoB sequences to demonstrate the significance of mooring sequence in RNA editing [14, 20], we converted chicken apoB sequences to rat apoB sequences and our data confirmed the importance and necessity of mooring sequences in RNA editing. The nucleotide changes in
748
Vol. 254, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Aligned nucleotide sequences of the apoB mRNA editing regions in five species. The numbering corresponds to the published human apoB100 sequence [23]. The chicken sequence is provided on the top line, and nucleotide differences from the chicken sequences in the other species compared are aligned beneath. Hyphens represent nucleotides identical to the chicken sequence. The nucleotides corresponding to apoB mRNA editing codon and mooring sequence are underlined. The 29-nucleotide conserved cassette (MC-29) is marked with a square. The efficiency elements (nts. 6609-6640 and 6717-6747) described by Hersberger and Innerarity [18] are marked with a square. Elements required for editing and editing enhancement (nts. 6615-6628 and 6726-6752) are highlighted with the shade.
this region during evolution have profound consequences on RNA editing. Our study demonstrates that neither the 11-nt mooring sequence, nor the entire 29-nt conserved cassette alone were sufficient for editing enhancement activity in chicken enterocyte S100 extracts. In order to elucidate the sequence element(s) required for in vitro editing enhancement activity, we engineered chimeric rat apoB/chicken apoB (pRC), or chicken apoB/rat apoB (pCR) constructs. Using these chimeric RNAs, rat enterocyte S100 extracts edited RC-RNA twice as efficiently as CR-RNA (28% vs. 15%), and in the presence of chicken enterocyte extracts, the proportion of edited RC-RNA, and CR-RNA were increased to ;77% and ;35%, respectively. Therefore, replacement of rat apoB RNA on either the 59 or 39 end of the 29-nt conserved cassette provides the sequence elements required for chicken editing enhancement activity. Using deletion strategy on pRC, we showed that nucleotides 6615-6629 were important for inducing editing and editing enhancement activity. Like the mooring sequence or the 29-nt conserved cassette, this region (nts. 6615-6628) is A/T rich (A 1 T 5 62%) and fully conserved across mammalian species (Fig. 4). Comparison of this stretch (nts. 6615-6628) to chicken sequences reveals 5 nucleotide substitutions. The significance of this region was also demonstrated by Hersberger and Innerarity [18], who showed that rabbit apoB nucleotides 6609-6629 contains important elements for editing at C-6666. By replacing rat apoB sequence (nucleotides 66906756) in the 39 of the 29-nt conserved cassette, chicken enterocyte S100 extracts also enhanced the proportion of editing of chimeric CR-RNA. Deletion studies revealed that nucleotides 6726-6752 provided sufficient information for apoB RNA editing and editing enhancement. Unlike the 59 upstream element, this region is much more diverged among
different vertebrate species, but it is A/T rich (A 1 T 5 73%) (Fig. 4). This may in part account for the finding that chicken enterocyte S100 extracts enhance rat, pig, and human RNAs with various editing efficiencies in vitro [5]. Interestingly, Hersberger and Innerarity also located a 31-nt fragment (nucleotides 6717-6747) which increased apoB RNA editing of C-6666 [18]. Sowden et al have proposed a “population gating” hypothesis to explain the apoB RNA editing mechanism by which cis- and transacting factor interact to determine editing efficiency [22]. The “gate” (edtosomes) evaluates each RNA passing through. Presumably, 59 element of nts. 6615-6628 and 39 element of nts. 6726-6752 provide the characteristics for increasing the efficiency of RNA editing and editing enhancement of C-6666. ACKNOWLEDGMENTS This research was supported by National Institute of Health Grants HL-53441 (to BBT) and HL-56668 (to LC) and American Heart Association Grant-In-Aid (to BBT). We are grateful to Dr. Kunihisa Kobayashi from Baylor College of Medicine for the preparation of computer graphics. We are also grateful to Dr. Eva Zsigmond from Institute of Molecular Medicine, University of TexasHouston for critical reading this manuscript.
REFERENCES
749
1. Teng, B. B., Burant, C. F., and Davidson, N. O. (1993) Science 260, 1816 –1819. 2. Navaratnam, N., Morrison, J. R., Bhattacharya, S., Patel, D., Funahashi, T., Giannon, F., Teng, B. B., Davidson, N. O., and Scott, J. (1993) J. Biol. Chem. 268, 20709 –20712. 3. Giannoni, F., Chou, S.-C., Skarosi, S. F., Verp, M. S., Field, F. J., Coleman, R. A., and Davidson, N. O. (1995) J. Lipid Res. 36, 1664 –1675. 4. Driscoll, D. M., and Zhang, Q. (1994) J. Biol. Chem. 269, 19843– 19847.
Vol. 254, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
5. Teng, B. B., and Davidson, N. O. (1992) J. Biol. Chem. 267, 21265–21272. 6. Yamanaka, S., Poksay, K. S., Balestra, M. E., Zeng, G.-Q., and Innerarity, T. L. (1994) J. Biol. Chem. 269, 21725–21734. 7. Schock, D., Kuo, S.-R., Steinburg, M. F., Bolognino, M., Sparks, J. D., Sparks, C., and Smith, H. C. (1996) Proc. Natl. Acad. Sci. USA 93, 1097–1102. 8. Yang, Y., Yang, Y., Kovalski, K., and Smith, H. C. (1997) J. Biol. Chem. 272, 27700 –27706. 9. Lau, P. P., Zhu, H.-J., Nakamuta, M., and Chan, L. (1997) J. Biol. Chem. 272, 1452–1455. 10. Driscoll, D. M., Wynne, J. K., Wallis, S. C., and Scott, J. (1989) Cell 58, 519 –525. 11. Teng, B. B., Black, D. D., and Davidson, N. O. (1990) Biochem. Biophys. Res. Commun. 173, 74 – 80. 12. Hodges, P., and Scott, J. (1992) Trends Biochem. Sci. 17, 77– 81. 13. Smith, H. C. (1993) Seminars in Cell Biol. 4, 267–278. 14. Backus, J. W., and Smith, H. C. (1991) Nuc. Acids Res. 19, 6781– 6786.
15. Backus, J. W., and Smith, H. (1992) Nuc. Acids Res. 20, 6007– 6014. 16. Backus, J. W., Schock, D., and Smith, H. C. (1994) Biochim. Biophys. Acta 1219, 1–14. 17. Driscoll, D. M., Lakhe-Reedy, S., Oleksa, L. M., and Martinez, D. (1993) Mol. Cellular Biol. 13, 7288 –7294. 18. Hersberger, M., and Innerarity, T. L. (1998) J. Biol. Chem. 273, 9435–9442. 19. Stevenson, S. C., Marshall-Neff, J., Teng, B. B., Lee, C. B., Roy, S., and McClelland, A. (1995) Arteriosclerosis & Thrombosis 15, 479 – 484. 20. Shah, R. R., Knott, T. J., Legros, J. E., Navaratnam, N., Greeve, J. C., and Scott, J. (1991) J. Biol. Chem. 266, 16301–16304. 21. Anant, S., MacGinnitie, A. J., and Davidson, N. O. (1995) J. Biol. Chem. 270, 14762–14767. 22. Sowden, M., Hamm, J. K., Spinelli, S., and Smith, H. C. (1996) RNA 2, 274 –288. 23. Knott, T. J., Wallis, S. C., Powell, L. M., Pease, R. J., Lusis, A. J., Blackhart, B., McCarthy, B. J., Mahley, R. W., Levy-Wilson, B., and Scott, J. (1986) Nucleic Acids Res. 14, 7501–7503.
750